Spores from single fronds of three different taxa of Pteridium aquilinum (L.) Kuhn were collected at different sites in Scotland, England and Sri Lanka. Gametophytes developed from these spores were treated to produce arrays of genetically identical clones. Sporophyte formation was determined when such clones from the same or different gametophytes, derived from the same frond, were combined in pairs in all possible ways to produce a diallel mating scheme. A recurring pattern of presence or absence of sporophyte formation indicated the occurrence of two genetic classes defined by no or very few sporophytes in pairs within either class but high frequency production in pair combinations of clones from different classes. The usual failure of sister clone pairs to produce a sporophyte contrasts with the frequently high incidence of sporophyte formation on the part of single, isolated non-cloned gametophytes. This conflict of evidence is discussed in relation to genetic incompatibility or, alternatively, the control of antheridia formation. The genetic differences revealed in cloned gametophytes provide an empirical way of determining whether a given stand of bracken is made up of more than one individual.